This investigation is designed to assess the impact of alternate combinations of intraoperative anesthesia and postoperative analgesia on the human stress response, selected morbidities, and the cost of medical care for patients undergoing major vascular surgery. Improvement in patient outcome and reduced utilization of medical resources have been suggested to result from using regional anesthetic techniques as compared to general anesthesia alone for large operative procedures. Studies that report this finding have suffered from methodologic problems including the use of non-equivalent modalities for postoperative pain relief and the failure to assure optimized clinical care in both groups. It remains unknown whether the benefits discovered are due to specific features of either the intraoperative or the postoperative technique, or to unrevealed aspects of patient management. To clearly separate the influence of time period and technique, remove physician bias, and provide comparable care in all groups we propose a five-year double blind clinical trial of 240 patients undergoing surgery of the abdominal aorta. Subjects will be randomized to one of four groups to receive either regional supplemented general anesthesia or general anesthesia alone during surgery, and either intravenous or epidural patient controlled analgesia after surgery. Specific aims of the study are to compare, between the four groups: 1) indices of the human stress response, 2) mortality and morbidity, and 3) return to physiologic independence and cost of medical care. All aspects of perioperative medical management will be specified by protocol. Data collection will include: 96 hours of continuous Holter monitoring; serial EKG's, and physical examinations; serial blood samples for cardiac isoenzymes, norepinephrine, epinephrine; urinary cortisols; vital sign analysis; discharge chart review and three and six month follow-up. Unique features of this proposal are the standardization of clinical care throughout the perioperative period, inclusion of all four combinations of intraoperative anesthesia and postoperative analgesia, patient and treating physicians blinded, and evaluation of the stress response. We hope to resolve issues of importance to both the quality and the cost of medical care and contribute to the understanding of physiologic factors which underlie perioperative morbidity. A finding of significant differences between groups will allow clinicians to adopt methods with due regard for both safety and economy while a finding of no difference will allow choice based on preference.